Neural-network-based adaptive hybrid-reflectance model for 3-D surface reconstruction

This paper proposes a novel neural-network-based adaptive hybrid-reflectance three-dimensional (3-D) surface reconstruction model. The neural network automatically combines the diffuse and specular components into a hybrid model. The proposed model considers the characteristics of each point and the variant albedo to prevent the reconstructed surface from being distorted. The neural network inputs are the pixel values of the two-dimensional images to be reconstructed. The normal vectors of the surface can then be obtained from the output of the neural network after supervised learning, where the illuminant direction does not have to be known in advance. Finally, the obtained normal vectors are applied to enforce integrability when reconstructing 3-D objects. Facial images and images of other general objects were used to test the proposed approach. The experimental results demonstrate that the proposed neural-network-based adaptive hybrid-reflectance model can be successfully applied to objects generally, and perform 3-D surface reconstruction better than some existing approaches.     

Multi-class object detection in tunnels from 3D point clouds: An auto-optimized lazy learning approach

This research develops an auto-optimized lazy learning approach named BOKNN (Bayesian optimized K nearest neighbor) method to detect seepage and multi-classify various objects (e.g., segment, pipe, track, support, and cable) in operating tunnels from 3D point clouds. Firstly, the 3D laser scanning is employed to acquire raw point cloud data, and the equidistant pooling for down-sampling is conducted to improve class imbalance issues and enhance the efficiency. Then, the K-nearest neighbor (KNN) model is built on the trimmed dataset, where the Bayesian optimization is performed to obtain the optimal combination of hyper-parameters in the KNN model. A realistic cross-river tunnel section in China is used as a case study to demonstrate the applicability and effectiveness of the developed approach. Results indicate that (1) The established BOKNN model displays a high performance in multi-class detection, together with a total accuracy of 0.935, a macro F1 score of 0.896, and a weighted F1 score of 0.939. (2) It performs well even in minor class detection, and the detection of seepage is conservative, where only 4.1% of seepage points are misclassified as non-seepage points. (3) It displays better detection performance than the other representative machine learning models (i.e, Adaboost, Support Vector Machine, and Naive Bayes). The developed approach is nonparametric and training-free, which can be used as a decision tool to substitute the present manual detection and improve the detection efficiency.     

Efficient and automatic plane detection approach for 3-D rock mass point clouds

The detection of planar regions from three-dimensional (3-D) laser scanning point clouds has become more and more significant in many scientific fields, including 3-D reconstruction, augmented reality and analysis of discontinuities. In rock engineering, planes extracted from rock mass point clouds are the foundational step to build 3-D numerical models of rock mass, which is significant in analysis of rock stability. In the past, several approaches have been proposed for detecting planes from TLS point clouds. However, these methods have difficulties in processing rock points because of the uniqueness of rock. This paper introduces a novel and efficient method for plane detection from 3-D rock mass point clouds. Firstly, after filtering the raw point clouds of rock mass acquired through laser scanning, the point cloud is split into some small voxels according to the specified resolution. Then, for the purpose of acquisition of high-quality growth units, an accurate coplanarity test process is used in each voxel. Meanwhile, the accurate neighborhood information can be built according to the result of coplanarity test. Finally, small voxels are clustered into a completed plane by region growing and the procedure of postprecessing. The performance of this method was tested in one icosahedron point cloud and three rock mass point clouds. Compared with the existing methods, the results demonstrate superior performance of our method in the field of plane detection.

     

Slice-based building facade reconstruction from 3D point clouds

ABSTRACT

We propose a comprehensive strategy to reconstruct urban building geometry from three-dimensional (3D) point clouds. First, the point clouds are segmented using a rough-detail segmentation algorithm, and refinements guided by topological relationships are performed to rectify the segmentation mistakes. Then, the semantic features (such as facades and windows) that belong to the buildings are recognized and extracted. Next, each facade is cut into a sequence of slices. The initial models are recovered by sequentially detecting and connecting the anchor points. Finally, due to the regular arrangements of windows, a template-matching method relying on the similarity and repetitiveness of the windows is proposed to recover the details on building facades. The experimental results demonstrate that our method can automatically reconstruct the building geometry and detailed window structures are better depicted.     

An evolutionary approach to the extraction of object construction trees from 3D point clouds

In order to extract a construction tree from a finite set of points sampled on the surface of an object, we present an evolutionary algorithm that evolves set-theoretic expressions made of primitives fitted to the input point-set and modeling operations. To keep relatively simple trees, we use a penalty term in the objective function optimized by the evolutionary algorithm. We show with experiments successes but also limitations of this approach.     

An adaptive normal estimation method for scanned point clouds with sharp features

Normal estimation is an essential task for scanned point clouds in various CAD/CAM applications. Many existing methods are unable to reliably estimate normals for points around sharp features since the neighborhood employed for the normal estimation would enclose points belonging to different surface patches across the sharp feature. To address this challenging issue, a robust normal estimation method is developed in order to effectively establish a proper neighborhood for each point in the scanned point cloud. In particular, for a point near sharp features, an anisotropic neighborhood is formed to only enclose neighboring points located on the same surface patch as the point. Neighboring points on the other surface patches are discarded. The developed method has been demonstrated to be robust towards noise and outliers in the scanned point cloud and capable of dealing with sparse point clouds. Some parameters are involved in the developed method. An automatic procedure is devised to adaptively evaluate the values of these parameters according to the varying local geometry. Numerous case studies using both synthetic and measured point cloud data have been carried out to compare the reliability and robustness of the proposed method against various existing methods.     

An approach to combine progressively captured point clouds for BIM update

Building information models (BIMs) provide opportunities to serve as an information repository to store and deliver as-built information. Since a building is not always constructed exactly as the design information specifies, there will be discrepancies between a BIM created in the design phase (called as-designed BIM) and the as-built conditions. Point clouds captured by laser scans can be used as a reference to update an as-designed BIM into an as-built BIM (i.e., the BIM that captures the as-built information). Occlusions and construction progress prevent a laser scan performed at a single point in time to capture a complete view of building components. Progressively scanning a building during the construction phase and combining the progressively captured point cloud data together can provide the geometric information missing in the point cloud data captured previously. However, combining all point cloud data will result in large file sizes and might not always guarantee additional building component information. This paper provides the details of an approach developed to help engineers decide on which progressively captured point cloud data to combine in order to get more geometric information and eliminate large file sizes due to redundant point clouds.     

Stratification approach for 3-D euclidean reconstruction of nonrigid objects from uncalibrated image sequences.

This paper addresses the problem of 3-D reconstruction of nonrigid objects from uncalibrated image sequences. Under the assumption of affine camera and that the nonrigid object is composed of a rigid part and a deformation part, we propose a stratification approach to recover the structure of nonrigid objects by first reconstructing the structure in affine space and then upgrading it to the Euclidean space. The novelty and main features of the method lies in several aspects. First, we propose a deformation weight constraint to the problem and prove the invariability between the recovered structure and shape bases under this constraint. The constraint was not observed by previous studies. Second, we propose a constrained power factorization algorithm to recover the deformation structure in affine space. The algorithm overcomes some limitations of a previous singular-value-decomposition-based method. It can even work with missing data in the tracking matrix. Third, we propose to separate the rigid features from the deformation ones in 3-D affine space, which makes the detection more accurate and robust. The stratification matrix is estimated from the rigid features, which may relax the influence of large tracking errors in the deformation part. Extensive experiments on synthetic data and real sequences validate the proposed method and show improvements over existing solutions.     

BIM reconstruction from 3D point clouds: A semantic registration approach based on multimodal optimization and architectural design knowledge

Reconstructing semantically rich building information model (BIM) from 2D images or 3D point clouds represents a research realm that is gaining increasing popularity in architecture, engineering, and construction. Researchers have found that architectural design knowledge, such as symmetry, planarity, parallelism, and orthogonality, can be utilized to improve the effectiveness of such BIM reconstruction. Following this line of enquiry, this paper aims to develop a novel semantic registration approach for complicated scenes with repetitive, irregular-shaped objects. The approach first formulates the architectural repetition as the multimodality in mathematics. Thus, the reconstruction of repetitive objects becomes a multimodal optimization (MMO) problem of registering BIM components which have accurate geometries and rich semantics. Then, the topological information about repetition and symmetry in the reconstructed BIM is recognized and regularized for BIM semantic enrichment. A university lecture hall case, consisting of 1.9 million noisy points of 293 chairs, was selected for an experiment to validate the proposed approach. Experimental results showed that a BIM was satisfactorily created (achieving about 90% precision and recall) automatically in 926.6 s; and an even more satisfactory BIM achieved 99.3% precision and 98.0% recall with detected semantic and topological information under the minimal effort of human intervention in 228.4 s. The multimodality model of repetitive objects, the repetition detection and regularization for BIM, and satisfactory reconstruction results in the presented approach can contribute to methodologies and practices in multiple disciplines related to BIM and smart city.     

Automatic segmentation of point clouds from multi-view reconstruction using graph-cut

In multi-view reconstruction systems, the recovered point cloud often consists of numerous unwanted background points. We propose a graph-cut based method for automatically segmenting point clouds from multi-view reconstruction. Based on the observation that the object of interest is likely to be central to the intended multi-view images, our method requires no user interaction except two roughly estimated parameters of objects covering in the central area of images. The proposed segmentation process is carried out in two steps: first, we build a weighted graph whose nodes represent points and edges that connect each point to its k-nearest neighbors. The potentials of each point being object and background are estimated according to distances between its projections in images and the corresponding image centers. The pairwise potentials between each point and its neighbors are computed according to their positions, colors and normals. Graph-cut optimization is then used to find the initial binary segmentation of object and background points. Second, to refine the initial segmentation, Gaussian mixture models (GMMs) are created from the color and density features of points in object and background classes, respectively. The potentials of each point being object and background are re-calculated based on the learned GMMs. The graph is updated and the segmentation of point clouds is improved by graph-cut optimization. The second step is iterated until convergence. Our method requires no manual labeling points and employs available information of point clouds from multi-view systems. We test the approach on real-world data generated by multi-view reconstruction systems.     

Three-dimensional surface reconstruction of tree canopy from lidar point clouds using a region-based level set method

In this article, a novel method is proposed for three-dimensional (3D) canopy surface reconstruction of trees using a region-based level set method. Both individual tree crowns and clusters of trees are first marked for further exploration. Multiple horizontal slices corresponding to different heights are obtained. The 3D structure of tree canopy is built using raw data from lidar point clouds. Also, new applications are proposed based on the new method for 3D forest reconstruction. The biomass parameters of the forest, including tree intersection area, tree equivalent crown radius, and canopy volume, can be calculated from stacking 2D slices of trees. Tree types are also identified and classified. The results indicate that this approach is effective for 3D surface reconstruction of forests including individual trees and clusters of trees, and that critical forest parameters (such as tree intersection area, tree position, and canopy volume) can be derived for the evaluation and measurement of biophysical parameters of forests.     

Statistical process monitoring approach for high-density point clouds

Statistical process control (SPC) methods have been extensively applied to monitor the quality performance of manufacturing processes to quickly detect and correct out-of-control conditions. As sensor and measurement technologies advance, there is a continual need to adapt and refine SPC methods to effectively and efficiently use these new data-sets. One of the most state-of-the-art dimensional measurement technologies currently being implemented in industry is the 3D laser scanner, which rapidly provides millions of data points to represent an entire manufactured part’s surface. Consequently, this data has a great potential to detect unexpected faults, i.e., faults that are not captured by measuring a small number of predefined dimensions. However, in order for this potential to be realized, SPC methods capable of handling these large data-sets need to be developed. This paper presents an approach to performing SPC using point clouds obtained through a 3D laser scanner. The proposed approach transforms high-dimensional point clouds into linear profiles through the use of Q–Q plots, which can be monitored by well established profile monitoring techniques. In this paper point clouds are simulated to determine the performance of the proposed approach under varying fault scenarios. In addition, experimental studies were performed to determine the effectiveness of the proposed approach using actual point cloud data. The results of these experiments show that the proposed approach can significantly improve the monitoring capabilities for manufacturing parts that are characterized by complex surface geometries.     

An evaluation of pose-normalization algorithms for point clouds introducing a novel histogram-based approach

Building Information Modeling is growing more relevant as digital models are not only used during the construction phase but also throughout the building’s life cycle. The digital representation of geometric, physical and functional properties enables new methods for planning, execution and operation. Digital models of existing buildings are commonly derived from surveying data such as laser scanning which needs to be processed either manually or automatically throughout various steps. Aligning point clouds along the coordinate system’s main axes (also commonly known as pose normalization) is a task benefitting any point cloud processing workflow, be it manual or automated. With the goal of automating this task, we compare various existing methods and present our own approach based on point density histograms. We conclude this paper by comparing and discussing all methods in terms of speed and robustness.     

大规模孔洞点云的快速重建算法研究 *

针对实际中经常存在的含有孔洞的点云数据 ,在原多层重建算法的基础上提出了一种可以进行点云补洞的快速曲面重建算法。首先对散乱点云数据进行空间自适应八叉剖分 ,然后对点云数据进行由粗到精的多层插值 ,建立隐式曲面方程 ,最后提出了两种加快重建的方法。加速算法可以减少重建时间 ,非常有利于处理大规模点云。实验结果证明 ,本算法对点云孔洞修补效果良好 ,重建速度快 ,效率高。     

Complete residential urban area reconstruction from dense aerial LiDAR point clouds

We present an automatic system to reconstruct 3D urban models for residential areas from aerial LiDAR scans. The key difference between downtown area modeling and residential area modeling is that the latter usually contains rich vegetation. Thus, we propose a robust classification algorithm that effectively classifies LiDAR points into trees, buildings, and ground. The classification algorithm adopts an energy minimization scheme based on the 2.5D characteristic of building structures: buildings are composed of opaque skyward roof surfaces and vertical walls, making the interior of building structures invisible to laser scans; in contrast, trees do not possess such characteristic and thus point samples can exist underneath tree crowns. Once the point cloud is successfully classified, our system reconstructs buildings and trees respectively, resulting in a hybrid model representing the 3D urban reality of residential areas.     

Splat-based surface reconstruction from defect-laden point sets

We introduce a method for surface reconstruction from point sets that is able to cope with noise and outliers. First, a splat-based representation is computed from the point set. A robust local 3D RANSAC-based procedure is used to filter the point set for outliers, then a local jet surface – a low-degree surface approximation – is fitted to the inliers. Second, we extract the reconstructed surface in the form of a surface triangle mesh through Delaunay refinement. The Delaunay refinement meshing approach requires computing intersections between line segment queries and the surface to be meshed. In the present case, intersection queries are solved from the set of splats through a 1D RANSAC procedure.     

3-D reconstruction from tomographic data using 2-D active contours.

Reconstructing three-dimensional (3-D) shapes of structures like internal organs from tomographic data is an important problem in medical imaging. Various forms of the deformable surface model have been proposed to tackle it, but they are either computationally expensive or limited to tubular shapes. In this paper a 3-D reconstruction mechanism that requires only 2-D deformations is proposed. Advantages of the proposed model include that it is conformable to any 3-D shape, efficient, and highly parallelizable. Most importantly, it requires from the user an initial 2-D contour on only one of the tomograph slices to start with. Experimental results are shown to illustrate the performance of the model.